Notices
993 Forum 1995-1998
Sponsored by:
Sponsored by:

Recent SAI fix that so far has not worked

Thread Tools
 
Search this Thread
 
Old 07-19-2007, 08:34 AM
  #1  
dhicks
Drifting
Thread Starter
 
dhicks's Avatar
 
Join Date: Oct 2004
Location: Rockville, MD
Posts: 2,483
Likes: 0
Received 0 Likes on 0 Posts
Default Recent SAI fix that so far has not worked

Rationale......if the air cannot get to the O2 sensors via the exhaust port, bypass the exhaust port and intoduce the air to the exhaust system as near to the exhaust port as possible. This should then satisfy the O2 sensors!

Remove air box. Detatch the rubber hose off the top of the SAI one way valve, leave this one way valve in place. Insert a new one way valve into the now open hose from the SAI air pump. From the underside of the new one way valve run a hose (I used Goodyear heater hose ID 1/2") out under where the air box will be, and out over the top of the fan housing. Cut hose to length so that it sits ontop of the fan housing but not stick out over it. Now insert an OD 1/2" "t" piece in the end of this tube. Replace air box.

Remove the heat exchangers (HE) and weld a nut (I used a Stainless Steel one) onto the top of flat piece just prior to where it connects to the tubing leading to the CATS....do to each HE, then drill hole in the heat HE exit pipe through the center of the nut. I then used a piece of stainless steel tubing, ID 1/4", OD 1/2" that was threaded to fit the nut and was about 1" proud when screwed into the nut. Obtain some aluminum tubing, about 12" with ID 1/2" and this should slide over the SS tube screwed into the HE. I used aluminum as I thought there would be less heat transfer up the length of the tube and it would be easier to bend.

If you now look up from under the car from where the HEs were towards where the light clusters are you will see two alumium discs which are just covers for two holes in the sheet metal, remove these two discs, they just pop out. Now remove the aluminum pipe from the SS adapter and replace the heat exchanger on the car. Now came some of the fiddle, you have to deliver the aluminum pipe through the hole in the sheet metal (the ones where you remove the aluminum discs from) to the SS adapter screwed into the nut in the HE....you may have to do some bending here but it is quite easy to do just a fiddle to get the right minor bends. Once you have done this you clamp the aluminum tube onto the SSadapter to make an airtight seal....I did this by using a little exhaust repair tape around the seal and then clamping.

On the top of the aluminum pipe attach more heater hose and clamp, run the hose up to the "t" then cut to length and attach to the "t". I then used black ties to hold the whole thing in place and wow it looks semi professional.......if only it had worked.....anybody have any idea why this did not work????
Old 07-19-2007, 09:46 AM
  #2  
jrb964
Instructor
 
jrb964's Avatar
 
Join Date: Oct 2001
Location: Arizona
Posts: 225
Likes: 0
Received 1 Like on 1 Post
Default

When you say "...if only it had worked..." what do you mean. Did you reset the ODE-II codes? How long afterwards did the CEL light return? Is the new code for banks 1-3 or 4-6?

My wrench and I are going to do basically the same thing next week. We're installing a new SAI valve, and then come off that with a "T" off the valve, and plug the hole that directed the gases through the heads. We then going to run use heat resistant tubing of 1/4" ID or less and run it along the metal trays on the side of the engine, and then near the back of th engine install a 90 degree exiting to the underside of the vehicle. From there are procedures are similar.

It may be that the ID of the tube is too large, thus limiting the pressure.
Old 07-19-2007, 10:09 AM
  #3  
dhicks
Drifting
Thread Starter
 
dhicks's Avatar
 
Join Date: Oct 2004
Location: Rockville, MD
Posts: 2,483
Likes: 0
Received 0 Likes on 0 Posts
Default

Yes I reset the codes and ran the car as per nornal, the codes for SAI reappeared as per normal, exactly as they did prior to the "fix", within 100 miles. Yes I wondered about the ID and its ability to delivery the air with sufficient pressure, let me know how yours goes. Where are you planing to egress into the exhaust system? What heat resistant tubing are you going to use?
Old 07-19-2007, 03:41 PM
  #4  
dhicks
Drifting
Thread Starter
 
dhicks's Avatar
 
Join Date: Oct 2004
Location: Rockville, MD
Posts: 2,483
Likes: 0
Received 0 Likes on 0 Posts
Default

Anybody else have any idea why this failed?
Old 07-19-2007, 04:11 PM
  #5  
Dan V
Race Car
 
Dan V's Avatar
 
Join Date: Jul 2005
Location: San Jose Kalifornia Demokratische Republik
Posts: 3,916
Likes: 0
Received 1 Like on 1 Post
Default

Seems there are 2 main variables to consider (FWIW. I'm not a trained engineer) .
1. The volume of air delivered by the new plumbing - is it reduced any significant amount?

2. The location of the air injection. Note where the original air injection ports are in relation to the exhaust valves -- how does this proximity impact the parameters that the OBDII is monitoring?
Old 07-19-2007, 04:26 PM
  #6  
dhicks
Drifting
Thread Starter
 
dhicks's Avatar
 
Join Date: Oct 2004
Location: Rockville, MD
Posts: 2,483
Likes: 0
Received 0 Likes on 0 Posts
Default

Dan
Thanks for the reply....yes I had thought about the volume....when I did the flush I looked at the holes that exit into the port and just sort of guessed that a 1/2" pipe would deliver about the same amount as all those together, maybe thats too big a pipe and I need to go to 1/4"....as to the proximity to the exhuast port itself, that is a good question....I just got as near as I could...the attached pic shows where I welded the nut for the injection of air.....that was about as close as I could get

Last edited by dhicks; 09-07-2013 at 05:14 PM.
Old 07-19-2007, 05:29 PM
  #7  
ppressle
Rennlist Member
 
ppressle's Avatar
 
Join Date: Aug 2004
Location: Los Altos, CA
Posts: 941
Likes: 0
Received 1 Like on 1 Post
Default

Has one of the steel plugs in the manifold possibly rusted through which is allowing the air to leak out?
Old 07-19-2007, 09:05 PM
  #8  
jrb964
Instructor
 
jrb964's Avatar
 
Join Date: Oct 2001
Location: Arizona
Posts: 225
Likes: 0
Received 1 Like on 1 Post
Default

David, I took your problem to my wrench and we discussed your situation. First, we think your pipe is too large. We're going to to use 1/8 or 3/16. Look at the passages in the heads, there between 1/8 and 3/16, we did measure them exactly. Secondly, did you replace you SAI valve. It could be defecitive.

Just some thoughts. As I said I'm doing mine on next Tuesday...I'll keep you informed.
Old 07-20-2007, 12:05 AM
  #9  
ppressle
Rennlist Member
 
ppressle's Avatar
 
Join Date: Aug 2004
Location: Los Altos, CA
Posts: 941
Likes: 0
Received 1 Like on 1 Post
Default

Why would too large a pipe be an issue? Same volume of air would flow through albiet at a lower velocity.
Old 07-20-2007, 12:10 AM
  #10  
cabrio993
Race Car
 
cabrio993's Avatar
 
Join Date: Feb 2005
Location: Johns Creek, GA
Posts: 4,682
Likes: 0
Received 10 Likes on 8 Posts
Default

Originally Posted by ppressle
Why would too large a pipe be an issue? Same volume of air would flow through albiet at a lower velocity.
Pressure would be different, it may not reach the cat with enough PPMs for the O2 sensors to detect the right quantity of oxigen in the air/co2 mix that is expecting, thus triggering the SAI cel.
Old 07-20-2007, 12:26 AM
  #11  
ppressle
Rennlist Member
 
ppressle's Avatar
 
Join Date: Aug 2004
Location: Los Altos, CA
Posts: 941
Likes: 0
Received 1 Like on 1 Post
Default

I guess you might need a little more pumping to get the larger volume pipe "up to pressure" to match the exhaust system pressure, but the pump should be putting out the correct pressure. Difference would be only the amount needed to pressurize the volume of the pipe. After that it should just be the same volume of air getting to the cat which would cause it to heat up at the same rate. yes?
Old 07-20-2007, 03:10 AM
  #12  
Mike J
Addict
Lifetime Rennlist
Member
 
Mike J's Avatar
 
Join Date: May 2001
Location: Vancouver, BC
Posts: 8,363
Received 71 Likes on 58 Posts
Default

That should work...I am not sure why. Do you have a OBD-II reader that can monitor the O2 voltages? You can tell if the pump is affecting the O2 sensors that way. You know you should see a lean condition on the O2 sensors when the pump is running.

I was playing with exactly the same idea so I am glad you have tried it. Have you considered using the O2 sensor test holes on the CAT as air injection spots? This way you don't have to weld in anything but it will requiire a bit of custom machining to make a custom fitting since these holes are fairly small (like M8...I did not measure them). They also might not be large enough to deliver the air needed.

Your pipe might be the wrong size. The passages in the cam carriers are really small and long so that would restrict the air quite a bit. You certainly can restrict the rubber hose with a clamp and try again. I doubt if the air pump is tuned that closely.

There is also a shut off valve before the pump (at least that is what I think it is). If you pull the hose off before your Al pipe can you feel the air being pumped to confirm that is working correctly?

Good work so far though...I was thinking this would be a cool product to sell to 993 owners except for the EPA liablities...

Cheers,

Mike
Old 07-20-2007, 06:01 AM
  #13  
Garth S
Rennlist Member
 
Garth S's Avatar
 
Join Date: Oct 2003
Location: Nova Scotia
Posts: 7,210
Likes: 0
Received 16 Likes on 13 Posts
Default

Too much air injected would be my guess as well - stick a (control) valve in the line, or pinch the tube off as Mike noted. The SAI pump is now running into less head/restriction than previously, so is delivering more mass .... and is likely drawing less amperage: I do not know if the latter is a measured parameter in the 993 ECU ... but it appears to be in a similarly balles-up SAI system I'm playing with on our Passat 1.8 Turbo.
Old 07-20-2007, 07:38 AM
  #14  
dhicks
Drifting
Thread Starter
 
dhicks's Avatar
 
Join Date: Oct 2004
Location: Rockville, MD
Posts: 2,483
Likes: 0
Received 0 Likes on 0 Posts
Default

Thanks for all the input guys....I will test with small ID tubing and see if that works....the pump is running, I can hear it. Only have an Actron reader that detects faults so cannot measure O2 readings....its so frustrating as I thought I had it licked!!!
Old 07-20-2007, 09:28 AM
  #15  
cabrio993
Race Car
 
cabrio993's Avatar
 
Join Date: Feb 2005
Location: Johns Creek, GA
Posts: 4,682
Likes: 0
Received 10 Likes on 8 Posts
Default

Originally Posted by dhicks
Thanks for all the input guys....I will test with small ID tubing and see if that works....the pump is running, I can hear it. Only have an Actron reader that detects faults so cannot measure O2 readings....its so frustrating as I thought I had it licked!!!
Baby steps...I think you're onto something.


Quick Reply: Recent SAI fix that so far has not worked



All times are GMT -3. The time now is 05:29 AM.